MAGScoT

MAGScoT (MAG Scoring Tool) scores and refines metagenome-assembled genomes (MAGs) from several binning tools. It estimates the completeness and contamination of each bin from GTDB (release 207) single-copy marker genes, merges overlapping bins from different tools, and selects the best set of non-overlapping bins.

MAGScoT

License

Free to use and open source under GNU GPLv3.

Available

Roihu-CPU: 1.1, via the bio-apps module.

Usage

MAGScoT is part of the bio-apps collection on Roihu. Load the bio-apps module tree and then the MAGScoT module:

module load bio-apps/v202603
module load magscot/1.1

MAGScoT is an R script. The module sets $MAGSCOT_INSTROOT to the installation directory, which contains the script and, under hmm, the GTDB marker gene profiles. The module also loads Prodigal and HMMER.

Find the marker genes in the contigs: predict the genes with Prodigal, search them for markers with HMMER and combine the hits into one table:

prodigal -p meta -i contigs.fa -a contigs.faa -o contigs.genes
hmmsearch -o tigr.out --tblout tigr.hits --noali --notextw --cut_nc --cpu 8 \
  $MAGSCOT_INSTROOT/hmm/gtdbtk_rel207_tigrfam.hmm contigs.faa
hmmsearch -o pfam.out --tblout pfam.hits --noali --notextw --cut_nc --cpu 8 \
  $MAGSCOT_INSTROOT/hmm/gtdbtk_rel207_Pfam-A.hmm contigs.faa
grep -v "^#" tigr.hits | awk '{print $1"\t"$3"\t"$5}' > contigs.hmm
grep -v "^#" pfam.hits | awk '{print $1"\t"$4"\t"$5}' >> contigs.hmm

MAGScoT reads the bins from a tab-separated file with three columns and no header: bin, contig and binning tool. Bin names must be unique across all the tools, so the commands below add the tool name to each bin name. For example, to combine the bins from SemiBin2, COMEBin, CONCOCT and MetaBAT2:

awk -F'\t' 'NR>1 && $2!=-1{print "semibin_"$2"\t"$1"\tsemibin2"}' semibin_out/contig_bins.tsv > contigs_to_bin.tsv
awk -F'\t' '{print "comebin_"$2"\t"$1"\tcomebin"}' comebin_out/comebin_res/comebin_res.tsv >> contigs_to_bin.tsv
awk -F',' 'NR>1{print "concoct_"$2"\t"$1"\tconcoct"}' concoct_output/clustering_merged.csv >> contigs_to_bin.tsv
awk '/^>/{b=FILENAME; sub(/.*\//,"",b); sub(/\.fa$/,"",b); print "metabat_"b"\t"substr($1,2)"\tmetabat2"}' bins/bin.[0-9]*.fa >> contigs_to_bin.tsv

By default, MetaBAT2 writes its bins only as FASTA files, so the last command reads the contig names from the numbered bin files (bins/bin.1.fa and so on) that metabat2 -o bins/bin writes.

Score and refine the bins:

Rscript $MAGSCOT_INSTROOT/MAGScoT.R -i contigs_to_bin.tsv --hmm contigs.hmm

MAGScoT writes the scores of the original bins to MAGScoT.scores.out, and the refined bins and their contigs to MAGScoT.refined.out and MAGScoT.refined.contig_to_bin.out. Change the MAGScoT prefix with -o. With bins from only one binning tool, MAGScoT only scores them.

Example batch script

#!/bin/bash
#SBATCH --job-name=magscot
#SBATCH --account=<project>
#SBATCH --output=output_%j.txt
#SBATCH --error=errors_%j.txt
#SBATCH --partition=small
#SBATCH --time=04:00:00
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=8
#SBATCH --mem=16G

module load bio-apps/v202603
module load magscot/1.1

prodigal -p meta -i contigs.fa -a contigs.faa -o contigs.genes
hmmsearch -o tigr.out --tblout tigr.hits --noali --notextw --cut_nc --cpu $SLURM_CPUS_PER_TASK \
  $MAGSCOT_INSTROOT/hmm/gtdbtk_rel207_tigrfam.hmm contigs.faa
hmmsearch -o pfam.out --tblout pfam.hits --noali --notextw --cut_nc --cpu $SLURM_CPUS_PER_TASK \
  $MAGSCOT_INSTROOT/hmm/gtdbtk_rel207_Pfam-A.hmm contigs.faa
grep -v "^#" tigr.hits | awk '{print $1"\t"$3"\t"$5}' > contigs.hmm
grep -v "^#" pfam.hits | awk '{print $1"\t"$4"\t"$5}' >> contigs.hmm

Rscript $MAGSCOT_INSTROOT/MAGScoT.R -i contigs_to_bin.tsv --hmm contigs.hmm

Replace <project> with your CSC project (for example project_2001234).

See creating a batch job script for Roihu for more information about running batch jobs.

Support

CSC Service Desk

More information

MAGScoT GitHub repository